University of Florida scientists want to assess livestock mobility more quickly and accurately, ultimately contributing to farm animal health and production.
To do this, they will use artificial intelligence (AI) to analyze high-definition video of the animals as they move.
Samantha Brooks, a UF/IFAS geneticist and associate professor of equine physiology — along with other UF researchers, has received a $49,713 grant from the Agricultural Genome to Phenome Initiative (AG2PI) for this study.
The team will combine machine learning with gait analytics to accelerate their assessment of livestock mobility.
Brooks gives an example of how this technology can help: In horses, one vet can do a basic examination of lameness in about 15 minutes.
“Our long-term goal is to build an automated pipeline that can provide near-real-time results just seconds after the animal passes the camera,” Brooks said. “This pilot project is a first step towards that goal.”
Brooks and her colleagues mainly work with horses because they are an excellent model for locomotion and because scientists can quickly collect a lot of data.
She and her lab are already working with some 2,000 video clips of horses in motion. Brooks thanks the hard work of graduate student Madelyn Smythe and the generosity of hundreds of horse owners in central Florida for the video.
“The large video library will make it possible to build accurate models to track the movements of the animals in the video frame,” Brooks said. “While we started with the horse, what we’re learning here will translate to similar models for other four-legged farm animals.”
For this project, they will also build AI models to analyze videos of cattle, pigs and small ruminants.
As they review the data, researchers will look at equine characteristics such as standing time, stride length and limb extension. In cattle and pigs, scientists are more interested in asymmetry and postures that indicate pain for abnormal function in one or more limbs.
Brooks said she wants to help other scientists and farm animal owners because AI, while helpful, isn’t always intuitive.
“Artificial intelligence approaches could accelerate our ability to measure complex movement characteristics in livestock, with better accuracy than the human eye,” Brooks said. “Yet AI tools are often not biologist-friendly, nor are they ready for challenging applications on the farm.”
“To address these challenges, we hope to adapt and merge existing AI methodologies into an analytics package that is accessible to scientists from different backgrounds and deployable in a variety of livestock management environments,” she said.
For example, the technology could detect lameness in livestock that pass a camera every day. Imagine dairy cattle entering the parlor, warning the farmer early on about potentially serious health problems, and with less effort from farm staff.
Funded by the U.S. Department of Agriculture’s National Institute of Food and Agriculture, AG2PI is a three-year project ending in 2023.
The goal of AG2PI is to connect crop and livestock scientists with each other and those working in data science, statistics, engineering and social sciences to identify shared problems and collaborate on solutions.